GLM-5: From Vibe Coding to Agentic Engineering

Re:GLM-5: From Vibe Coding to Agentic Engineering by monooso on Wednesday February 11, 2026 @02:22PM (#46979511)

This aligns very closely with my experience.

When left to its own devices, GLM-4.7 frequently tries to build the world. It's also less capable at figuring out stumbling blocks on its own without spiralling.

For small, well-defined tasks, it's broadly comparable to Sonnet.

Given how incredibly cheap it is, it's useful even as a secondary model.

[ Parent | on HN ]

Re:GLM-5: From Vibe Coding to Agentic Engineering by antoniuschan99 on Wednesday February 11, 2026 @08:48PM (#46983904)

How is the web search functionality? I have only used deepseek to lower costs from gpt api but had to incorporate a serper to actually do web searches

[ Parent | on HN ]

Re:GLM-5: From Vibe Coding to Agentic Engineering by monooso on Thursday February 12, 2026 @06:40AM (#46987525)

Assuming you've configured the web search MCP [1], it works well.

[1]: https://docs.z.ai/devpack/mcp/search-mcp-server

[ Parent | on HN ]